profile
viewpoint

defunctzombie/ansible-coreos-bootstrap 231

[NOT MAINTAINED] bootstrap a coreos machine for control via ansible

browserify/node-util 155

node.js util module as a module

defunctzombie/badginator 115

badges for your badges

browserify/browser-resolve 95

resolve function which support the browser field in package.json

defunctzombie/bitcoin-address 61

bitcoin address verification and other tools

defunctzombie/changelog 50

changelog release management tool

defunctzombie/away 32

monitor when a user is inactive on a page

defunctzombie/bamboo 22

javascript model library for basic CRUD

defunctzombie/chain-js 8

callback chaining library for javascript

defunctzombie/browser-stacks 6

stackframes from various browsers

issue commenttendermint/tendermint

Regression with chain catchup behavior introduced in commit 1000378

@marbar3778 any updates on gaia release?

defunctzombie

comment created time in 3 days

issue commenttendermint/tendermint

Regression with chain catchup behavior introduced in commit 1000378

Awesome! Do you know if gaia will be updated with this new release or is it something I should request?

defunctzombie

comment created time in 10 days

issue commenttendermint/tendermint

Regression with chain catchup behavior introduced in commit 1000378

So far it looks like the nodes are keeping up with each other as I would expect.

defunctzombie

comment created time in 11 days

issue commenttendermint/tendermint

Regression with chain catchup behavior introduced in commit 1000378

I've deployed a patched gaiad - I'll report back in 12 hrs and see if the nodes are all keeping up with the chain as I would expect - it takes some time for them to fall out of sync since after a restart they all catch up to the chain head first before continuing other actions.

defunctzombie

comment created time in 12 days

issue commenttendermint/tendermint

Regression with chain catchup behavior introduced in commit 1000378

I’ll give it a try today. Apologies for the delay in responding.

On Mon, Aug 3, 2020 at 3:02 AM Erik Grinaker notifications@github.com wrote:

Hi @defunctzombie https://github.com/defunctzombie. Any chance you could try the fix out today? We're preparing to cut 0.33.7 as well, and would like to include this fix if we can get it confirmed.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/tendermint/tendermint/issues/5112#issuecomment-667933570, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAUWOCYFOHMFCHCBAZKRH3R62DMNANCNFSM4OYBFOPA .

defunctzombie

comment created time in 12 days

issue commentoutline/outline

[feature request] Collection Document/Page

I'd say this request was to enhance that UX to make it editable similar to a full page.

defunctzombie

comment created time in 18 days

issue commenttendermint/tendermint

Regression with chain catchup behavior introduced in commit 1000378

What about a config option to disable?

I still don't think the retry logic as introduced in the bugfix is appropriate for how the Cosmos chain operates so being able to disable it would allow our nodes to stay updated with the chain.

defunctzombie

comment created time in 21 days

issue commentZondax/ledger-filecoin

Add screen to display address on the device.

@jleni Were in the app is this? I've installed v0.17.2 on a test device and clicked through the app's menus. I did not see any settings or screen to show the address.

defunctzombie

comment created time in 23 days

issue openedZondax/ledger-filecoin

Add screen to display address on the device.

A nice security enhancement is to display the Filecoin address on the device (rather than relying on a connected UI). This provides an additional verification around the user's account address.

Happy to clarify more if this post is not descriptive enough.

created time in 23 days

issue commenttendermint/tendermint

Regression with chain catchup behavior introduced in commit 1000378

A backport would be 👍 since the bugfix that was applied to the v0.32 family is what caused the specific regression - so getting v0.32 back to a similar state for chain syncing would help operators with remote signers that are connected to several front nodes.

defunctzombie

comment created time in 24 days

issue commenttendermint/tendermint

Regression with chain catchup behavior introduced in commit 1000378

What's the best way to give #5140 a try? Change the go.mod file in gaia? Will the version of tendermint that change is based on work with the latest stable gaia?

I'd have a stronger preference for a config option to just disable the retry functionality altogether (ideally default no retry and allow folks who feel strongly that they need to tune retry to turn it on). I prefer simpler connection flow logic over pre-tuned retries especially on a fast moving chain.

Some thoughts on the earlier comments:

a) proposer might miss a chance to create a block and receive a reward

To me that seems like it is on the validator to ensure their connections are reliable to avoid missing the reward. The reward is also not that outsized that missing it as a proposer is significantly detrimental. Maybe the answer here is to have 1 immediate retry - if you failed that too then I still believe the right move is to move on.

b) validator might miss a chance to precommit for a block and get slashed for not signing

Missing one precommit won't get you slashed. If you are disconnected and come back online - the node would catch up and you would start signing current blocks which would prevent you from being slashed. You would need to be disconnected for ~16 hrs before liveliness slashing hits - if you are down that long - this retry logic would actually do more to negatively impact your ability to come back online and be at the head of the chain.

If you want to help folks avoid liveliness slashing - that should be done at a higher design/protocol level rather than per single block.

defunctzombie

comment created time in 25 days

pull request commentoutline/outline

chore: Remove env variables in webpack bundle

Absolutely - you know best what makes sense for the project. I wanted to provide the Dockerfile as an option since I had noted earlier that I would.

On the PR itself - something I've found helpful is to consolidate the config into a single file (config.js, env.js, etc) and use that in various locations rather than "window.env". It allows for a single place to document and simpler management of how the environment configuration is loaded (today from window, maybe tomorrow elsewhere) and avoid the window global sprinkled around the project files. Since you are already touching many of these locations now might be a nice time to make such a change.

tommoor

comment created time in a month

pull request commentoutline/outline

chore: Remove env variables in webpack bundle

Sorry I hadn't gotten around to making a PR for the referenced docker image PR - but if you wanted to keep your env vars during the build you can accomplish this using docker build ARGS as seen here:

https://github.com/Polychain/outline/blob/v0.43.3-pc0/Dockerfile

tommoor

comment created time in a month

issue openedtendermint/tendermint

Regression with node catchup behavior introduced in #1000378

We are running Cosmos gaiad 2.0.11 which uses v0.32.11 of the tendermint sdk. This version of the tendermint sdk introduced a bugfix via https://github.com/tendermint/tendermint/pull/4709/commits/100037877080bf73c951531732363451d51b5236 which is causing some of our nodes to fall behind in chain state.

We use the remote signing feature of a nodes to connect several nodes to a single backing validator (a scheme that has worked excellently for us since cosmoshub launch and we made available via our open-source validator here: https://gitlab.com/polychainlabs/tendermint-validator). With this approach the remote validator maintains a high water mark and will reject signatures requests for heights below the watermark. Prior to the change in https://github.com/tendermint/tendermint/pull/4709/commits/100037877080bf73c951531732363451d51b5236 our nodes would all keep up with the blockchain equally.

After the change - if a node happens to fall behind by 1 block (due to having lower block propagation time or network connectivity - it will continue to fall behind and fall out of sync with the chain and never catch up.

Given how fast new blocks are produced on the network - we see retrying as rarely useful and certainly not useful with such a long timeout.

My suggestion is to revert the commit and allow the next block to attempt a new signature and move forward.

created time in a month

pull request commentoutline/outline

run "yarn build" when making the production docker image

@tommoor still interested in landing a version of this? I have some new changes to the approach (via using docker build args) that works more seamlessly with injecting env vars (for webpack) to the build process during the docker build.

defunctzombie

comment created time in a month

create barnchPolychain/outline

branch : v0.43.3-pc0

created branch time in 2 months

startedmarcello3d/lascaux-sketch

started time in 2 months

issue closedbrowserify/node-browser-resolve

Maintenance

Hi @defunctzombie, this module is heavily used by browserify and we'd like to do some updates (like bumping the resolve dependency). I understand you're not actively working on this package and I'd be happy to maintain it for the foreseeable future. Would you be open to transfering it to the @browserify org? Thanks!

closed time in 2 months

goto-bus-stop

issue commentbrowserify/node-browser-resolve

Maintenance

I've transferred the repo to the browserify org and granted goto-bus-stop owner access to the browser-resolve package. Thank you for picking up maintenance of this module.

goto-bus-stop

comment created time in 2 months

issue commentdefunctzombie/node-browser-resolve

Maintenance

Sure - what do I need to do to make that happen?

goto-bus-stop

comment created time in 2 months

push eventPolychain/amber-network-entities

Roman Shtylman

commit sha d20a37252f16e619532aa764ada5e9168dccec04

entities: add Polychain Labs entity package

view details

push time in 2 months

push eventPolychain/amber-network-entities

Roman Shtylman

commit sha 2e582e77a981f5ea6f45da5f5d9c20274f2f7458

entities: add Polychain Labs entity package

view details

push time in 2 months

push eventPolychain/cosmos-sdk

Roman Shtylman

commit sha 9ace6df543291e29ca4b70dc9a80db6579a55a6f

add distribution keeper allocation hooks

view details

push time in 2 months

create barnchPolychain/core

branch : accounting-v0.3.5

created branch time in 2 months

create barnchPolychain/cosmos-sdk

branch : accounting-v0.37.11

created branch time in 2 months

create barnchPolychain/tendermint

branch : accounting-v0.32.10

created branch time in 2 months

more